JPH01190078A - Method for setting reference white level of picture reading device - Google Patents

Abstract

PURPOSE:To constantly maintain a white level in the vicinity of a reference white level by reading a white reference line by the line image sensor of plural bits constitution, comparing with a reference white level set beforehand for each bit and controlling the gain of a variable gain amplifier. CONSTITUTION:The reading of a white reference line 20 is executed by a line image sensor 13 prior to the reading of an original. White level data from the line image sensor 13, after being amplified by a variable gain amplifier 21, is inputted to an A/D converter 15 and compared with a reference white level V + ref for each bit. When the white level data are comparatively higher than the reference white level V + ref, the number of overflown bits is calculated by a first counter circuit 22, and when it reaches a previously set value, a count down is executed by an up and down counter 24. Consequently, the gain of the variable gain amplifier 21 is reduced and controlled, the level of the white level data to be inputted to the A/D converter 15 is lowered and approaches the reference white level V + ref.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for setting a reference white level in an image reading device such as a facsimile machine.

[Prior Art] FIG. 4 is a block diagram showing a schematic configuration of an image reading system in a conventional facsimile machine. A line image sensor 1 with a multi-bit configuration that scans and photoelectrically converts the focused light by a lens 12 that focuses the light.
3, a fixed gain amplifier 14 that amplifies the image signal output from the line image sensor 13 with a constant gain, and an amplified image that is between a preset reference white level V+ref and a reference black level V-rer. A/D (analog/digital) converter 15 that multivalue quantizes the signal
The signal quantized by the A/D converter 15 is sent to the output terminal V as an image reading signal.
Output from out.

In the A/D converter 15, the reference white level V+ver matches the amplified image signal level (read white level) when the white portion of the document 10 is read by the line image sensor 13, and the reference black level V -re
Image signal level after amplification when f reads the black part (
When the read black level matches the black level), the A/D converter 15
Since all the resolutions of are used, highly accurate multi-level quantization is possible. Therefore, in conventional image reading devices of this type, the gain of the fixed gain amplifier 14 and the A/D
The converter 15 was designed so that values satisfying the above conditions were selected as the reference white level V+ref and the reference black level V-ref.

[Problem 2A to be Solved by the Invention] However, the brightness of the light source 11 generally changes depending on the temperature during use, and the photoelectric conversion output level of the line image sensor 13 changes with this change in brightness. The photoelectric conversion output level also changes depending on the temperature characteristics of the line image sensor itself. Furthermore, the fixed gain amplifier 14 has temperature drift, and its characteristics change depending on the operating temperature. Therefore, due to these various causes, the A/D converter 15
The image signal input to the A/D converter 15 is easily affected by the operating temperature, and in particular, the read white level changes greatly depending on the operating temperature, and this change in white level may reduce the accuracy of multi-value quantization in the A/D converter 15. was there.

SUMMARY OF THE INVENTION Therefore, the present invention aims to provide a reference white level setting method for an image reading apparatus that can compensate for changes in the read white level due to operating temperature, etc., and can always maintain the white level near the reference white level. .

[Means and operations for solving communication problem] The reference white level setting method of the image reading device of the present invention reads a white reference line with a line image sensor having a multi-bit configuration, and changes this read data as white level data. After amplification with a gain amplifier, compare each bit with a preset reference white level to count the number of bits that exceed the reference white level, and when this count reaches a predetermined number, reduce the gain of the variable gain amplifier. The white level when the line image sensor reads the white reference line is controlled to be close to the reference white level.

Further, the reference white level setting method of the image reading device of the present invention is such that a white reference line is read by a line image sensor having a multi-bit configuration, the read data is amplified as white level data by a variable gain amplifier, and then the white level is set in advance. The number of bits missing from the reference white level is counted by comparing each bit with the reference white level, and when this count reaches a predetermined value, the gain of the variable gain amplifier is increased and the line image sensor becomes white. The self-level when reading the reference line is brought closer to the reference white level.

[Embodiment] An embodiment in which the method of the present invention is applied to an image reading system of a facsimile machine will be described below.

FIG. 1 is a block diagram of the image reading system of this embodiment, in which light sources 11. Lens 12. Line image sensor 13 with multiple bit configuration. A for multi-value quantization of an image signal between the reference white level V+rer and the reference black level V-rer
Since the /D converter 15 is similar to the conventional one shown in FIG. 4, the same reference numerals will be given and detailed explanation will be omitted. In FIG. 1, reference numeral 20 indicates a white reference line set by a white reference plate provided in the document reading section or a white reference portion (white portion corresponding to the first few lines) of the document 10 placed on the document reading section. , 21 is a variable gain amplifier that amplifies the image signal output from the line image sensor 13 and outputs it to the A/D converter 15. Then, in the A/D converter 15, the read white level amplified by the variable gain amplifier 21 and the reference white level V+rer are compared bit by bit, and overflow bits exceeding the reference white level V÷ver are detected by a first counter. The output signal is output to the circuit 22, and a plurality of upper bits lacking in the reference white level V-rcf are output to the area selection circuit 23.

The first counter circuit 22 counts the number of input overflow bits in one line, and when it reaches a preset count value, connects it to a clock (CK) terminal of an up/down counter 24 for up/down judgment. This circuit outputs pulse signals to the circuit 25 and the circuit 25, respectively. Here, since there is some variation between bits in one line in the line image sensor 13, the appropriate count value for the first counter circuit 22 is normally about 10% of the number of bits in one line. .

As shown in FIG.
Among the plurality of high-order bits inputted from the converter 15, the bits included in the selection area M indicated by diagonal lines in the figure are selected and outputted to the second counter circuit 26. In addition,
In FIG. 2, curves A, B, and C represent the A/D converter 1
5 shows changes in the self-level of the upper bits outputted from No. 5 at different operating temperatures. Here, the reason for providing the non-selected region H above the selected region M is to prevent the level of the bit located in the non-selected region H from exceeding 100% when the gain of the variable gain amplifier 21 is increased. This is for the purpose of

Further, the non-selected area below the selected area M is the black level area of the bit. Therefore, it is appropriate for the normal selection area M to be 38% to 75% or 38% to 80%.

The second counter circuit 26 counts the number of bits selected by the area selection circuit 23 in one line, and when a preset count value is reached, the up/down counter 2
4 clock (CK) terminal and up/down determination circuit 25
It outputs a pulse signal for each. Here, as the set count value of the second counter circuit 26, since there is some variation between each bit in the line image sensor 13 as described above, it is usually appropriate to set the count value to about 6026, which is the number of bits for one line. be. In addition, the above-mentioned No. 1.
The second counters 22, 26 each receive a reset signal R3.
If you input , the counting operation will stop.

The up/down determination circuit 25 includes a first counter circuit 22
When a pulse signal is input from the up/down counter 2
Outputs a signal to perform a down counter for 4,
When a pulse signal from the second counter circuit 26 is inputted, a signal is outputted that causes the up/down counter 24 to perform up-counting. Therefore, in the up/down counter 24, the countdown is performed when a pulse signal is generated from the first counter circuit 22, or when the pulse signal is generated from the first counter circuit 22.
When a pulse signal is generated from the counter circuit 26, counting up is performed at most once in one line.

Then, an output i corresponding to the count value from the up/down counter 24 is given to the variable gain amplifier 21 as a gain control signal for the variable gain amplifier 21.

FIG. 3 is a circuit diagram of the iI variable gain amplifier 21. As shown in FIG. In the figure, 31 is a decoder that decodes the 4-bit output QA-QD from the up-down counter 24 to generate a Cl-C1B signal, and 32 is a fixed device that amplifies the image signal output from the line image sensor 13 with a constant gain. Gain amplifier, 5WI-8WI (i is a group of switches, one of which is turned on by the output of the decoder 31, R1
-R18 is a voltage dividing resistor group that divides the output of the amplifier 32, and when SWl is turned on by the output from the decoder 31, the image signal from the line image sensor 13 is amplified with the smallest gain and the A/D converter 15 When 5W1G is turned on, the image signal is amplified with the largest gain and output to the A/D converter 15.

Next, the operation of this embodiment will be explained. The white reference line 20 is read by the line image sensor 13, and the read own level data is amplified by the variable gain amplifier 21.
/D converter 15. Note that variable gain amplifier 2
1, it is assumed that a substantially intermediate gain (for example, SW& is on) is selected as the gain corresponding to the initial value of the up/down counter 24.

The A/D converter 15 which receives the image signal of the white reference line 20 compares this white level data bit by bit with the reference white level V+rer, and determines the reference white level V+.
The bits exceeding rcl' are output to the first counter circuit 22. Further, a plurality of appropriate upper bits are output to the area selection circuit 23.

The first counter circuit 22 counts the number of bits exceeding the reference white level V + rcl' in one line.

When a preset count value is reached, pulse No. 18 is output to the up/down counter 24 and up/down determination circuit 25. Then, the up/down counter 24 counts down and outputs QA-
QD changes, and as the output changes due to this countdown, the output Cl-C1B of the decoder 31 is switched and controlled in a direction in which the gain of the variable gain amplifier 21 decreases (for example, SW7 is turned on). As a result, the amplification factor of the variable gain amplifier 21 decreases, and the level of the white level data input to the A/D converter 15 decreases.

On the other hand, the area selection circuit 23 selects, from among the plurality of high-order bits, the bits that are short of the reference white level V + ver and are included in the selection area M, and output them to the second counter circuit 26 . 1. The second counter circuit 26 counts the selected number of bits, and when a preset count is reached, a pulse signal is output to the up/down counter 24 and up/down determination circuit 25. Then, the up/down counter 24 counts up and the output QA-QD changes, and as the output changes due to this count-up, the output Cl-C of the decoder 31
1B is switched and controlled in a direction in which the gain of the variable gain amplifier 21 increases (for example, SW9 is turned on). As a result, the amplification factor of the variable gain amplifier 21 increases, and the A/D converter 1
The level of the own level data input to 5 increases.

In this embodiment, which operates as described above, first, the white reference line 20 is read by the line image sensor 13 prior to reading the document.

Then, the white level data from the line image sensor 13 is amplified by the variable gain amplifier 21 and then inputted to the A/D converter 15, and the reference white level V +re is set for each bit.
It is compared with l'. Here, when the white level data is relatively high compared to the reference white level V+ref, the number of overflowed bits is counted by the counter circuit 22 of '1s1, and when this counted value reaches a preset value, The up/down counter 24 counts down. As a result, the gain of the variable gain amplifier 21 is controlled to decrease, and the level of the white level data input to the A/D converter 15 is lowered to reduce the reference white level V◆ra.
It will approach t.

On the other hand, when the white level data is relatively low compared to the reference white level V+rar, the number of bits selected by the area selection circuit 23 is counted by the second counter circuit 26,
When this count reaches a preset value, the up/down counter 24 counts up. As a result, the gain of the variable gain amplifier 21 is controlled to increase, and the level of the white level data input to the A/D converter 15 increases to approach the reference white level V + rcr.

As described above, according to this embodiment, by controlling the gain of the variable gain amplifier 21 based on the data read from the white reference line 20 by the line image sensor 13, the level read from the line image sensor 13 can be adjusted. It is possible to approach the reference white level V +ver in the A/D converter 15. Therefore, before reading the original, the preset white reference line 20 is read several times (for example, once).
By repeating this process (approximately 0 times), the read white level at the operating temperature can be brought sufficiently close to the reference white level V+rai', the resolution of the A/D converter 15 can be increased, and highly accurate multi-value quantization is possible. becomes. In this state, the first. Reset signal R3 to the second counter circuit 22.26
Since the variable gain amplifier 21 can be fixed at the optimum gain by giving a value of Output terminal v out
is output to. Therefore, it is possible to always compensate for changes in the read white level due to the influence of operating temperature and maintain the white level close to the reference white level, and it is possible to always achieve stable, high-viscosity image reading.

In the above embodiment, the case where both the number of bits exceeding the reference white level and the number of bits missing are counted, but even if either one of the number of bits is counted, the A/D converter will 15 accuracy can be improved. Further, in the above embodiment, the method of the present invention is applied to an image reading system of a facsimile machine, but the present invention is not limited to this, and can also be applied to, for example, a copying machine.

[Effects of the Invention] As described in detail above, according to the present invention, it is possible to compensate for changes in the read white level due to operating temperature, etc., and to maintain the white level near the reference white level at all times, thereby improving the performance of the A/D converter. It is possible to provide a reference white level setting method for a color image reading device that improves accuracy.

[Brief explanation of the drawing]

Figures 1 to 3 are diagrams showing an embodiment in which the method of the present invention is applied to an image @ reading system of a facsimile machine, in which Figure 1 is a block diagram of the image reading system, and Figure 2 is an area selection diagram. FIG. 3 is a circuit configuration diagram of an n1 variable gain amplifier, and FIG. 4 is a block diagram showing an image reading system in a conventional facsimile machine. 13... Line image sensor, 15... A/D converter, 20... White reference line, 21... Variable gain amplifier, 22.26... First. Second counter circuit, 23... Area selection circuit, 24... Up/down counter, 25... Up/down determination circuit, 31...
·decoder. Applicant's agent Patent attorney Takehiko Suzue

Claims (2)

[Claims] (1) A white reference line is read by a line image sensor with a multi-bit configuration, this read data is amplified by a variable gain amplifier as white level data, and then compared bit by bit with a preset reference white level. counting the number of bits exceeding the white level, and when the counted value reaches a predetermined number, controlling the gain of the variable gain amplifier to decrease;
A method for setting a reference white level for an image reading device, characterized in that the white level when the line image sensor reads a white reference line is brought closer to the reference white level. (2) A white reference line is read by a line image sensor with a multi-bit configuration, the read data is amplified as white level data by a variable gain amplifier, and then compared bit by bit with a preset reference white level. The number of bits lacking in the white level is counted, and when this count reaches a predetermined number, the gain of the variable gain amplifier is increased and the white level is determined when the line image sensor reads the white reference line. A method for setting a reference white level for an image reading device, characterized in that the reference white level is set close to the reference white level.